A four-year project is proposed to develop and implement a randomized trial to determine whether the regular practice of breast self-examination (BSE) will reduce mortality from breast cancer. Following a one-year planning period, approximately 300,000 current or former female employees of the Shanghai Textile Industry Bureau (STIB) between the ages of 30 and 64 in 455 factories will be randomly allocated factory to an intervention or control group. Women will be interviewed to obtain information on risk factors for breast cancer and breast screening and self-examination experience. Those in the intervention group will be given intensive instruction in BSE using a combination of techniques. One year later, the instruction will be reinforced, and competency and compliance in BSE assessed in all women in the intervention group. Reminders to practice BSE will be given in the interim, and subsequently. During the fourth year of the study, all women in both groups will be actively followed up for development of breast cancer and to ascertain vital status; and a sample in each group will be selected to ascertain the level of screening for breast cancer (if any) in the control group, and the level of compliance and competency in practicing BSE in the intervention group. All data will be sent to Seattle for processing, editing, monitoring, and analysis. During the fourth year, an application will be written to fund follow-up and monitoring for compliance and competency for an additional eight years (10 years from initial intervention). This study has a power of approximately 80% to detect a reduction in breast cancer mortality of about 30% in 10 years. Interim analyses will be performed and the trial will be terminated early if either a significant reduction in breast cancer mortality is demonstrated before the trial is scheduled to end, or if the estimated difference in the level of screening between the two groups falls below 80% and subsequent investigation indicates that the situation cannot be improved. The STIB offers virtually optimal conditions for this trial and success is highly likely. The results will provide an estimate of the upper limits of the beneficial effect that can reasonably to expected of BSE in most situations, and will therefore provide information needed to estimate the potential impact that BSE could have in reducing mortality from cancer in the United States. This study could be classified as Phase 4 cancer control research in a defined population.